Assessing the Susceptibility of Cruciferous Lepidoptera to Cry1Ba2 and Cry1Ca4 for Future Transgenic Cruciferous Vegetables

Transcription

1 HORTICULTURAL ENTOMOLOGY Assessing the Susceptibility of Cruciferous Lepidoptera to Cry1Ba2 and Cry1Ca4 for Future Transgenic Cruciferous Vegetables A. M. SHELTON, 1,2 G. T. GUJAR, 3 M. CHEN, 1 A. RAUF, 4 R. SRINIVASAN, 5 V. KALIA, 3 A. MITTAL, 3 A. KUMARI, 3 K. RAMESH, 3 R. BORKAKATTI, 3 J. Z. ZHAO, 1,6 N. ENDERSBY, 7 D. RUSSELL, 7,8 Y. D. WU, 9 AND B. UIJTEWAAL 10 J. Econ. Entomol. 102(6): 2217Ð2223 (2009) ABSTRACT Advances in transgenic plants expressing Bacillus thuringiensis (Bt) insecticidal gene(s) offer a promising alternative to traditional insecticides for control of lepidopteran pests on important cruciferous vegetable crops such as cabbage and caulißower. A public-private partnership, the Collaboration on Insect Management for Brassicas in Asia and Africa (CIMBAA), was formed in 2005 with the goal of developing dual-gene Bt caulißower and cabbage, initially for India, to replace the use of broad spectrum, traditional insecticides. As a Þrst step in this effort, the major lepidopteran pests of cruciferous vegetable crops [Plutella xylostella (L.), Pieris rapae (L.), Pieris brassicae (L.), Crocidolomia binotalis (L.), Hellula undalis (F.), Diacrisia obliqua Walker, Spodoptera litura F., and Helicoverpa armigera (Hübner)] were collected over a large geographic area (India, Indonesia, Taiwan, China, Australia, and the United States) and tested against puriþed Cry1Ba2 and Cry1Ca4 toxins, the toxins proposed to be expressed in the CIMBAA plants. Our results demonstrate that Cry1Ba2 and Cry1Ca4 were effective against the primary target of the CIMBAA plants, P. xylostella, regardless of geographic location, and had LC 50 values 1.3 ppm. Furthermore, one or both toxins were effective against the other major pest Lepidoptera, except for S. litura or H. armigera which were less susceptible. No cross-resistance has been found between Cry1Ba2 and Cry1Ca4, suggesting cry1ba2 cry1ca4 caulißower and cabbage could be an effective and sustainable tool to control, P. xylostella, the key lepidopteran pest on cruciferous vegetable crops, as well as most other Lepidoptera. As the CIMBAA plants are being developed, further tests are needed to determine whether they will express these proteins at sufþcient levels to control all the Lepidoptera. Sustainable use of the dual-gene plants also is discussed. KEY WORDS Plutella xylostella, diamondback moth, Bacillus thuringiensis, Cry toxins 1 Department of Entomology, Cornell University/NYSAES, Geneva, NY Corresponding author, ams5@cornell.edu. 3 Division of Entomology, Indian Agricultural Research Institute, New Delhi , India. 4 Department of Plant Protection, Bogor Agricultural University, Indonesia. 5 AVRDC-The World Vegetable Center, Taiwan Current address: Pioneer Hi-Bred International Inc., 7300 NW 62nd Ave., P.O. Box 552. Johnston, IA Department of Genetics, University of Melbourne, Parkville, VIC 3010 Australia. 8 Natural Resources Institute, University of Greenwich, Chatham, Kent ME4 4TB, United Kingdom. 9, Department of Entomology, Nanjing Agricultural University, Nanjing , China. 10 Nunhems BV, Haelen, The Netherlands. The advent of transgenic plants expressing insecticidal protein(s) (Cry toxins) from the bacterium, Bacillus thuringiensis (Bt), has launched a new era in crop protection. In 2008, Bt plants were grown on 46.0 million ha in 22 countries (James 2008). The majority of this total was Bt maize expressing cry1ab or cry1f genes to protect it against stalk-boring and ear-infesting lepidopteran pests. Bt maize, Zea mays L., was grown on 31.6 million ha in 17 countries in Bt cotton expressing cry1ac, cry2ab2, and cry1f (in China, the Cowpea trypsin inhibitor [CpTI] is also used) was grown on 14.5 million ha in 10 countries in Use of Bt plants has resulted in positive changes in insecticide practices, farm income and environmental impact. Worldwide, Brookes and Barfoot (2008) estimated that between 1996 and 2006 the deployment of Bt cotton led to a 22.9% reduction of insecticides, provided economic beneþt of $9.6 billion, and reduced the environmental impact quotient (EIQ, a measure of a pesticideõs harm to the environment, Kovach et al. 1992) by 24.6%. During this same period, use of Bt maize led to a 5.0% reduction of insecticides, provided economic beneþt of $3.6 B and reduced the EIQ by 5.3%. Advances in insect-resistant transgenic crops in recent years also have offered a promising alternative to traditional synthetic insecticides for control of lepidopteran pests of cruciferous vegetables (Shelton et al. 2008). The two most widely grown cruciferous /09/2217Ð2223$04.00/ Entomological Society of America

2 2218 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 102, no. 6 vegetables are cabbage (Brassica oleracea ssp. capitata) and caulißower (Brassica oleracea ssp. botrytis). In 2005, the area of caulißower and cabbages harvested worldwide was 4.12 million ha (FAOSTAT 2007) and, of this total, 80% was grown in developing countries. Caulißower and cabbages are important vegetable cash crops for low-income farmers throughout Asia, Africa, Latin America, and the Caribbean. They serve as important staple dietary items and are high in folate, vitamins B and C and other micronutrients (HSPH 2007). India is the largest producer of caulißower and the second largest producer of cabbage in the world, behind China. Lepidopteran pests of cruciferous vegetables are severe economic threats to production worldwide. The diamondback moth, Plutella xylostella (L.) (Lepidoptera: Plutellidae), is considered the most destructive insect pest, especially in resource-poor regions, and now occurs wherever crucifers are grown and causes losses to the world economy of an estimated US$1 billion yearly (Talekar and Shelton 1993). Losses of caulißower and cabbage due to P. xylostella frequently reach 90% without the use of insecticides (CIMBAA 2008) and, even with frequent use of insecticides, substantial losses occur and threaten food security. In tropical areas where pest pressure is high, it is not uncommon to apply insecticides every other day. Such intense use of insecticides poses hazards to farmers, consumers and the environment and has caused populations of this insect to become resistant to most of the major insecticides (Shelton et al. 2008). Studies have shown excellent control of P. xylostella by B. oleracea plants carrying a synthetic or modiþed Bt gene (Metz et al. 1995, Cao et al. 1999, Jin et al. 2000, Bhattacharya et al. 2002). Transgenic collards with cry1ac or cry1c genes showed excellent control of susceptible P. xylostella larvae (Cao et al. 2005). Additional studies with a cry1c gene expressed in broccoli demonstrated excellent control of Cry1Ac-resistant P. xylostella (Cao et al. 1999), and studies with pyramided cry1ac and cry1c broccoli plants demonstrated excellent control of both Cry1C-resistant and Cry1Ac-resistant P. xylostella (Cao et al. 2002). Other Lepidoptera also can be problematic in various parts of the world and these include Pieris rapae (L.), Pieris brassicae (L.) (both Lepidoptera: Pieridae), Crocidolomia binotalis ( pavonana) (L.) (Lepidoptera: Pyralidae), Hellula undalis (F.) (Lepidoptera: Pyralidae), Diacrisia obliqua Walker (Lepidoptera: Arctiidae), Spodoptera litura F. (Lepidoptera: Noctuidae), and Helicoverpa armigera (Hübner) (Lepidoptera: Noctuidae). If Bt cruciferous vegetables are grown in certain areas, they may also be exposed to these Lepidoptera. Thus, it is important to assess the susceptibility of these pests to candidate proteins that are under consideration to be expressed in Bt crucifers. The Collaboration on Insect Management for Brassicas in Asia and Africa (CIMBAA 2008) is a publicprivate partnership focusing on developing dual (pyramided) Bt gene cabbage and caulißower initially for India and then for other parts of the developing world. Because some populations of P. xylostella are reported to have increased tolerance to Cry1A toxins in sections of India (Mittal et al. 2007) and other countries (Shelton et al. 2007) due to the use of Bt spray formulations, we believe it is wise to consider alternative toxins for expression in Bt plants. Based on previous research (Zhao et al. 2001, Mohan and Gujar 2002, Mittal et al. 2007), it was suggested that CIMBAA plants use cry1ca4 and cry1ba2 genes because the proteins expressed by these genes had good activity against P. xylostella and cross-resistance between the two toxins was not detected in this insect (Zhao et al. 2001). Furthermore, studies had shown that resistance to Cry1C in P. xylostella is polygenic (Zhao et al. 2000), making it more difþcult for the insect to evolve resistance to this protein. Finally, the private company (Nunhems BV, Haelen, The Netherlands) involved in CIMBAA has access to these two genes and could readily use them in its breeding program. In this article, we evaluated the insecticidal activity of puriþed Cry1Ba2 and Cry1Ca4 to strains of eight lepidopteran pest species of cruciferous vegetables collected from India, Indonesia, Taiwan, China, Australia, and the United States. Materials and Methods Collection, Rearing, and Maintenance of Insects. Insects were collected and bioassayed in each country using an agreed common protocol. The participating laboratories and the respective leaders were as follows: Indian Agricultural Research Institute, India (Gujar); Bogor Agricultural University, Indonesia (Rauf); AVRDC-The World Vegetable Center, Taiwan (Srinivasan); Nanjing Agricultural University, China (Wu); University of Melbourne, Australia (Russell); and Cornell University, USA (Shelton). In total, 26 populations of P. xylostella were collected in 2006Ð2007 as larvae from cabbage and caulißower Þelds from geographically different locations in India, Indonesia, Taiwan, and the United States. Sample sizes for all populations were 100 individuals. The laboratory population (Geneva 88) was used as a standard susceptible population, and four other laboratory populations also were included as references, with one of the populations being resistant to Cry1Ac. Fewer populations of the other Lepidoptera (P. rapae, P. brassicae, H. undalis, D. obliqua, S. litura, and H. armigera) were collected from Þelds in these countries as well as in China and Australia. Once collected, insects were transferred to the respective laboratories where larvae were reared on fresh cruciferous leaves until pupation. Moths were provided with a carbohydrate solution (honey or sugar water) and allowed to mate and lay eggs, and the emerging larvae were fed on crucifer leaves until sufþcient numbers of insects could be assayed in the earliest possible generation. B. thuringiensis Cry toxins. Cry1Ba2 and Cry1Ca4 puriþed toxins were obtained from R. Akhurst, Commonwealth ScientiÞc and Industrial Research Organization (Commonwealth ScientiÞc and Industrial Research Organization), Canberra, Australia. They

3 December 2009 SHELTON ET AL.: CRUCIFEROUS LEPIDOPTERA SUSCEPTIBILITY TO CRY1BA2 AND CRY1CA Table 1. Toxicity of Cry1Ba2 to field collected and laboratory strains of P. xylostella larvae in leaf-dip assay at 72-h reading Insect strain Country of collection Gen (F) n Slope (SE) LC 50 (95% FL) ppm 2 (df) Field Aligarh India F (0.19) 0.13 (0.07Ð0.23) 17.9 (4) Almora India F (0.26) 0.02 (0.01Ð0.03) 6.63 (3) Bangalore India F (0.20) 0.01 (0.003Ð0.02) 3.25 (3) Bangalore (PDBC) India F (0.22) 0.21 (0.13Ð0.49) 1.04 (4) Bojoura-1 India F (0.35) 0.15 (0.10Ð0.20) 3.62 (3) Bojoura-2 India F (0.27) 0.46 (0.22Ð8.14) 1.11 (3) Delhi (Najafgarh) India F (0.24) 0.01 (0.007Ð0.02) 5.15 (3) Delhi (Sarai Kale-khan-1) India F (0.26) 0.05 (0.03Ð0.08) 1.39 (3) Delhi (Sarai Kale-khan-2) India F (0.37) 0.02 (0.007Ð (2) Delhi (Sarai Kale-khan-3) India F (0.19) 0.03 (0.01Ð0.58) 0.54 (3) Hisar India F (0.21) 0.01 (0.007Ð0.02) 1.0 (3) Hyderabad India F (0.15) 0.06 (0.02Ð0.13) 5.04 (4) Jorhat India F (0.24) 0.11 (0.08Ð0.15) 16.4 (4) Katrain India F (0.18) 0.09 (0.05Ð0.21) 2.04 (3) Pune India F (0.23) 0.09 (0.06Ð0.14) 1.14 (3) Ranchi India F (0.18) 0.04 (0.02Ð0.05) 9.42 (3) Sirsa India F (0.43) 0.08 (0.06Ð0.11) 2.77 (4) Vadodara India F (0.27) 0.05 (0.02Ð0.14) 1.47 (3) Batu Indonesia F (0.37) 0.46 (0.21Ð0.69) 6.45 (3) Bedugul Indonesia F (0.64) 0.91 (0.57Ð1.43) 6.67 (3) Lembang Indonesia F (0.32) 0.40 (0.14Ð0.62) 5.17 (3) Malino Indonesia F (0.27) 0.86 (0.50Ð1.96) 8.96 (3) Wonosobo Indonesia F (0.32) 0.62 (0.50Ð0.74) 1.91 (3) Tainan Taiwan F (0.19) 0.65 (0.56Ð0.74) 7.13 (3) Kula, HI USA F (0.28) 0.24 (0.19Ð0.29) 1.28 (3) Panama, FL USA F (0.29) 0.23 (0.18Ð0.30) 1.45 (3) Laboratory Geneva 88 (lab standard) USA F (0.30) 0.43 (0.28Ð0.62) 3.07 (3) SZ a China F (0.35) 0.27 (0.13Ð0.45) 8.35 (3) SZ Cry1Ac-resistant a China F (0.65) 0.35 (0.22Ð0.50) 7.72 (3) Waite Australia F (0.21) 1.28 (1.05Ð1.61) 0.24 (4) Queensland Australia F (0.24) 0.24 (0.21Ð0.29) 0.12 (8) a The 120-h reading. were produced in Escherichia coli and trypsin activated and used in all the bioassays reported. Bioassays. The majority of tested Þeld populations were assayed in F1Ð3, although a few populations required additional rearing until sufþcient numbers of larvae could be obtained. Toxicity of Cry1Ba2 and Cry1Ca4 was tested against second instars of P. xylostella using a leaf-dip feeding assay (Zhao et al. 2001, Mittal et al. 2007), and similar procedures were conducted for larvae of the other species. Caulißower or cabbage leaves were cut into discs and dipped in solutions of 5Ð7 concentrations prepared with individual Cry toxins. The concentrations were calculated in ppm. Leaf discs ( 5 per concentration replicate) were dipped in the respective concentrations for 5Ð10 s, air-dried and placed individually into small plastic containers. From Þve to 10 larvae were released onto each disc and mortality was recorded at 72 h for P. xylostella (assays times varied for other species) and analyzed to calculate median lethal concentrations (LC 50 ). An entire bioassay for one population with all replicates, including the control, was done on the same day and experiments with control mortality 10% were discarded and repeated. The mortality data were analyzed using probit analysis (Ross 1977, Russell et al. 1977), and the LC 50 values and relevant statistics were calculated. Differences in toxicity were considered signiþcant when 95% Þducial limits (FL) of LC 50 values did not overlap. Results Cry1Ba2 was toxic to larvae of P. xylostella with LC 50 (ppm) values ranging from 0.01 (Bangalore, Delhi [Najafgarh] and Hisar) to 0.91 (Bedugul) among the 26 Þeld populations tested (Table 1). There were signiþcant differences in susceptibility between some populations, based on nonoverlapping 95% FL values of the LC 50. Of the Indian populations, the most tolerant population (Bojoura-2) was 46-fold more tolerant than the Bangalore, Delhi (Najafgarh) and Hisar populations. In Indonesia, the highest LC 50 value was 0.91 (Bedugul) and the lowest 0.40 ppm (Lembang), only a 2.3-fold difference. The LC 50 values for populations collected in the United States and Taiwan ranged between 0.23 and 0.65 ppm. The standard labsusceptible population (Geneva 88) had a value of 0.43 ppm. To get an idea of how this susceptible population compared with the other populations regardless of country, its LC 50 was 43-fold higher than the least susceptible populations (Bangalore, Delhi [Najafgarh] and Hisar) and 2.1-fold less than the most tolerant population (Bedugul). Cry1Ca4 was also toxic to larvae of P. xylostella with LC 50 (ppm) values ranging from 0.01 (Bangalore and Aligarh) to 1.18 (Tainan) among the 20 Þeld populations tested (Table 2). There were signiþcant differences in susceptibility between some populations, based on nonoverlapping 95% FL values of the LC 50.

4 2220 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 102, no. 6 Table 2. Toxicity of Cry1Ca4 to field-collected and laboratory strains of P. xylostella larvae in leaf-dip assay at 72-h reading Insect strain Country of collection Gen (F) n Slope (SE) LC 50 (95% FL) ppm 2 (df) Field Aligarh India F (0.15) 0.01 (0.01Ð0.02) 5.44 (5) Almora India F (0.18) 0.02 (0.01Ð0.04) (4) Bangalore India F (0.17) 0.01 (0.003Ð0.02) 7.32 (4) Bangalore (PDBC-Þeld) India F (0.32) 0.43 (0.27Ð1.07) 1.82 (4) Bojoura-1 India F (0.65) 0.19 (0.19Ð0.79) 6.35 (4) Bojoura-2 India F (0.17) 0.61 (0.25Ð0.10) 1.02 (4) Delhi India F (0.33) 0.10 (0.07Ð0.15) 9.17 (3) Delhi (Najafgarh) India F (0.26) 0.10 (0.07Ð0.15) 0.53 (4) Delhi (Sarai Kale-khan-3) India F (0.16) 0.06 (0.04Ð0.11) 0.82 (4) Katrain India F (0.64) 0.27 (0.19Ð0.42) 2.00 (3) Pune India F (0.18) 0.07 (0.04Ð0.10) 2.48 (4) Ranchi India F (0.25) 0.18 (0.13Ð0.28) 3.11 (4) Sirsa India F (0.26) 0.11 (0.08Ð0.16) 4.13 (4) Batu Indonesia F (0.44) 0.07 (0.01Ð0.16) 0.89 (3) Bedugul Indonesia F (0.22) 0.54 (0.29Ð1.04) 6.28 (3) Lembang Indonesia F (0.27) 0.44 (0.14Ð0.73) 4.81 (3) Malino Indonesia F (0.23) 0.41 (0.25Ð0.64) (3) Wonosobo Indonesia F (0.28) 0.49 (0.19Ð0.81) 3.41 (3) Tainan Taiwan F (0.13) 1.18 (0.93Ð1.55) 4.33 (3) Kula, HI USA F (0.31) 0.20 (0.15Ð0.26) 0.05 (3) Laboratory Geneva 88 (lab standard) USA F (0.20) 0.18 (0.14Ð0.22) 2.96 (4) SZ a China F (0.25) 0.22 (0.15Ð0.32) 5.61 (4) SZ Cry1Ac-resistant a China F (0.39) 0.43 (0.36Ð0.50) 0.13(3) Waite Australia F (0.17) 1.05 (0.71Ð1.37) 0.69 (3) Queensland Australia F (0.22) 0.19 (0.16Ð0.24) 0.95 (8) a The 120-h reading. Of the Indian populations, the most tolerant population (Bojoura-2) was 61-fold more tolerant than the Bangalore and Aligarh populations. The four populations from Indonesia had LC 50 values between 0.07 and 0.54, 7.7-fold difference, whereas the Taiwan population (Tainan) had an LC 50 value 2.2-fold higher than the highest value for an Indonesian population. The standard lab-susceptible population (Geneva 88) had a value of Compared with the other populations regardless of country, its LC 50 was 18-fold higher than the least susceptible populations (Bangalore) and 6.5-fold less than the most tolerant population (Tainan). All populations of P. rapae, P. brassicae, C. binotalis, H. undalis and D. obliqua were susceptible to Cry1Ba2 (Table 3). Susceptibility to Cry1Ba2 varied from 2.09 ppm (H. undalis from Taiwan) to 0.07 (C. binotalis from Delhi), a difference of 30-fold. Within a species, when multiple populations were tested the highest variability for Cry1Ba2 was 13-fold for C. binotalis. P. brassicae, C. binotalis, H. undalis, and D. obliqua were all susceptible to Cry1Ca4 with the highest LC 50 value of 2.68 ppm for the single population of P. brassicae (Table 4). The value for the single population of P. rapae was 19.22, 7.2-fold higher than the value for P. brassicae and 16.3-fold higher than the least susceptible population of P. xylostella (Tainan, Table 2). Compared with P. xylostella and some other species tested, H. armigera and S. litura were less susceptible to either toxin (data not shown in tables). Cry1Ba2 had low efþcacy against a S. litura strain collected from Indonesia (20% mortality at a concentration of 12.5 ppm) and Taiwan (26% mortality at a concentration of 5 ppm). It had similar efþcacy against two strains of H. Table 3. Toxicity of Cry1Ba2 to field collected lepidopteran larvae in leaf-dip assay Insect species (generation) Strain/country Assay time (h) n Slope (SE) LC 50 (95% FL) ppm 2 (df) Pieris rapae (10) Geneva/USA (0.37) 0.12 (0.10Ð0.15) 1.03 (3) P. rapae (1) Tainan/Taiwan (0.20) 0.17 (0.15Ð0.19) 0.57 (3) Pieris brassicae (1) Delhi/India (0.47) 0.12 (0.11Ð0.14) 5.12 (5) Crocidolomia binotalis (3) Delhi/India (0.18) 0.07 (0.04Ð0.11) (4) C. binotalis (3) Batu/Indonesia (0.48) 0.66 (0.59Ð0.73) 2.97 (3) C. binotalis (3) Cianjur/Indonesia (0.29) 0.97 (0.56Ð2.06) (4) C. binotalis (3) Serang/Indonesia (0.39) 0.91 (0.53Ð1.77) (3) C. binotalis (3) Wonosobo/Indonesia (0.31) 0.58 (0.33Ð0.87) 6.08 (3) C. binotalis (1) Tainan/Taiwan (0.14) 1.07 (0.90Ð1.27) 6.27 (3) Hellula undalis (3) Serang/Indonesia (0.56) 0.49 (0.37Ð0.58) 1.73 (3) H. undalis (1) Tainan/Taiwan (0.16) 2.09 (1.80Ð2.43) 1.08 (3) Diacrisia obliqua (1) Delhi/India (0.13) 1.82 (0.39Ð12.4) 7.68 (5)

5 December 2009 SHELTON ET AL.: CRUCIFEROUS LEPIDOPTERA SUSCEPTIBILITY TO CRY1BA2 AND CRY1CA Table 4. Toxicity of Cry1Ca4 to field collected lepidopteran larvae in leaf-dip assay Insect species (generation) Strain/country Assay time (h) n Slope (SE) LC 50 (95% FL) ppm 2 (df) Pieris rapae (10) Geneva/USA (0.32) (10.6Ð75.0) 5.10 (3) Pieris brassicae (1) Delhi/India (0.31) 2.68 (1.74Ð3.86) 6.23 (4) Crocidolomia binotalis (3) Delhi/India (0.21) 1.89 (0.12Ð0.34) 0.94 (4) C. binotalis (3) Cianjur/Indonesia (0.18) 0.88 (0.52Ð1.91) 9.95 (4) C. binotalis (3) Batu/Indonesia (0.37) 0.83 (0.55Ð1.24) 8.02 (3) C. binotalis (3) Serang/Indonesia (0.41) 1.58 (1.07Ð2.54) (3) C. binotalis (3) Wonosobo/Indonesia (0.31) 0.42 (0.32Ð0.52) 0.96 (3) C. binotalis (1) Tainan/Taiwan (0.20) 1.88 (0.91Ð3.68) (3) Hellula undalis (3) Serang/Indonesia (0.24) 0.59 (0.44Ð0.75) 0.24 (3) H. undalis (1) Tainan/Taiwan (0.15) 1.37 (1.17Ð1.63) 2.61 (3) Diacrisia obliqua (1) Delhi/India (0.10) 1.82 (0.52Ð15.3) 3.79 (7) armigera collected from China (no mortality at a concentration of 10 ppm), a population from India (no mortality at a concentration of 5 ppm) and Australia (3.3% mortality at a concentration of 10 ppm). Similarly, Cry1Ca4 did not show good efþcacy against a S. litura population from India (30% mortality at a concentration of 5 ppm), Indonesia (20% mortality at a concentration of 12.5 ppm) or Taiwan (34% mortality at a concentration of 5 ppm) or against a H. armigera population from China (no mortality at a concentration of 10 ppm). Likewise, it was less effective against a H. armigera population from India (no mortality at a concentration of 5 ppm) and Australia (3.3% mortality at a concentration of 10 ppm). Discussion This was the Þrst area-wide survey of susceptibility in P. xylostella to Cry1Ba2 in India or any other country we are aware of and the results show its potential for use in Bt crucifer crops. Although there were large and signiþcant variations in susceptibility to Cry1Ba2 between populations, the highest LC 50 value of a Þeld colony (Bedugul) was only 2.1-fold the value of the susceptible Geneva 88 population (0.43 ppm) and less than the Waite, another laboratory colony (1.28 ppm). Of the 26 Þeld populations tested, 22 had LC 50 values 0.43 ppm and should be considered very susceptible. Furthermore, we consider this range in susceptibility to be natural variation because we do not believe that Indian populations of P. xylostella have been exposed to Cry1Ba2. Although Cry1B is found in B. thuringiensis subsp. thuringiensis HD-2, no product based on this Bt strain is registered in India and we are not aware of any unofþcial use of this product. Although there were signiþcant differences between Þeld and laboratory populations in susceptibility to Cry1Ca4, the highest Þeld value was 0.54 ppm (Bedugul), which was only three-fold higher than the susceptible Geneva 88 population (0.18 ppm) and less than the Waite, another laboratory colony (1.47 ppm), also tested at 72 h. Of the 20 Þeld populations tested, 11 had LC 50 values 0.18 and should be considered very susceptible. In the current study both Cry1Ba2 and Cry1Ca4 were toxic to P. xylostella, but neither protein was consistently more toxic to all populations (Tables 1 and 2). In a previous study, Cry1B was found to be approximately Þve-fold more toxic than Cry1C on a single population of P. xylostella (Zhao et al. 2001). Regardless, our data suggest that CIMBAA plants expressing Cry1Ba2 and Cry1Ca4 could be effective against P. xylostella strains with different geographic backgrounds from India and other countries. We have already obtained such expression levels in broccoli plants developed at Cornell expressing Cry1C (Cao et al. 2002), but it should be pointed out that expression levels in the CIMBAA plants would also have to meet the requirements of high expression strategy to be effective in an insecticide resistance management (IRM) program (Bates et al. 2005). It is also worth noting that a population from China (SZ Cry1Acresistant), which was 2,000-fold resistant compared with SZ (Y.D.W., unpublished data), was not signiþcantly more or less susceptible to either Cry1Ba2 or Cry1Ca4 compared with the SZ population. This again conþrms the lack of cross-resistance between these three proteins. Variation in susceptibility in geographic populations and its relationship to control on Bt plants has been the focus of considerable attention. Macrae et al. (2005) reported that transgenic soybean with highdose expression of Cry1A provided full control of laboratory and Þeld-collected strains of Anticarsia gemmatalis Hübner (Lepidoptera: Noctuidae) with different geographic backgrounds (from the United States or Argentina). Likewise, Wu et al. (1999) found a 100-fold variation in LC 50 values of Chinese populations of H. armigera to the Cry1Ac being introduced in Bt cotton, Gossypium hirsutum L., at the time, but even now control by Bt cotton remains effective (Ferré et al. 2008). Furthermore, even with high levels of resistance (as determined by LC 50 values in laboratory bioassays), studies have shown little if any survivorship on high expressing Bt plants for Ostrinia nubilalis (Hübner), (e.g., Li et al. 2007). This latter fact suggests that Bt plants may overcome large variations in susceptibility that are based on laboratory assays. Furthermore, it also suggests that other assays, such as growth inhibition, may be more appropriate for assessing resistance to the actual Bt plant (Anilkumar et al. 2008). However, conducting surveys using LC 50 values to particular Bt proteins is appropriate at least to determine a pestõs susceptibility.

6 2222 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 102, no. 6 The current study indicates that Cry1Ba2 and Cry1Ca4 seem to be appropriate proteins to be expressed by CIMBAA plants for India and other countries where the main lepidopteran pest is P. xylostella. This is fortunate because there is strong interest in producing Bt crucifers for India (Srinivasan et al. 2005), and such plants could become a major component in an overall integrated pest management (IPM) program (Shelton et al. 2008). Insecticide use on cabbages and caulißower in the Kulla Valley of India, a major production area, constitutes the highest part of production costs (12.5%), and the majority of growers (53.3%) spray on a calendar basis and their choice of insecticide is usually a broad-spectrum insecticide with a high environmental impact quotient (Badenes- Perez and Shelton 2006). Similarly, S. Sandur (unpublished data) found the costs of insecticides and their application on brassicas in Karnataka, India, to be 25% of the total production costs, with up to 35% of the crop lost to insect pests annually even where insecticides were used weekly. He estimated the total costs of insecticide and its application on cabbage in India to be US$54 million per year. However, based on the results presented here, it must be emphasized that other key Lepidoptera also will need to be controlled and that S. litura and H. armigera are less susceptible to Cry1Ba2 and Cry1Ca4 than is P. xylostella. Despite our laboratory results on these two species, the Þnal evidence must be obtained when the transgenic events are created and bioassays are conducted with the plant tissue. It is possible that expression levels could be high enough to control both species, especially because the more susceptible neonates (rather than the second instars we tested) would encounter the toxins when they emerge from eggs laid on the plant. However, careful evaluation of alternative strategies for these two species and other nonlepidopteran pests is required for the development of the overall IPM program for Bt crucifers. The evolution of resistance to Bt plants is a prime concern for their sustainability and various strategies have been developed and implemented to reduce the likelihood of resistance development (Gould 1998, Bates et al. 2005, Ferré et al. 2008). Since their introduction in 1996, there is only one veriþed case of an insect that has developed resistance to a Bt plant in the Þeld (S. frugiperda on Bt maize in Puerto Rico; Ferré et al. 2008) and that was in a crop expressing a single Bt gene (Cry1F). However, because of the past history of Þeld resistance evolution in P. xylostella to all major classes of insecticides, including Cry toxins in some parts of the world (Shelton et al. 2007), it is essential that Bt crucifers be introduced in a full setting of IPM practices. A major component of this is to introduce the plants with a strong IRM strategy. For P. xylostella it has already been demonstrated that using dual gene plants that use Bt genes that are not cross-resistant is a more durable strategy (Zhao et al. 2003). Other IPM strategies are also important and include cultural practices such as plowing under crop residues, breaking the pest cycle by having host-free periods and using selective insecticides or other control measures which minimize adverse impacts on natural enemies of P. xylostella (Hoy et al. 2007). Numerous studies have demonstrated the importance of natural enemies, especially parasitoids, in helping to manage P. xylostella (Kirk and Bordat 2004). Recent studies have shown that Bt crucifers that express high levels of Cry1C are not harmful to an important parasitoid of P. xylostella, Diadegma insulare (Cresson), and this was in stark contrast to commonly used synthetic and organic insecticides (Chen et al. 2008). Ongoing studies (M.C., unpublished) have also conþrmed that Cry1Ca4 is not toxic to several predators of P. xylostella. Thus, Bt crucifers have the potential to control P. xylostella while conserving its natural enemies, which also may be important for reducing the evolution of resistance to Bt crucifers in P. xylostella (Gould 1998). Acknowledgments This work was supported in part by a grant from USAID through the International Food Policy Research Institute and the Program for Biosafety Systems. The work conducted by AVRDC was funded by the Eiselen Foundation. References Cited Anilkumar, K. J., A. Rodrigo-Simón, J. Ferré, M. Pusztai- Carey, S. Sivasupramaniam, and W. J. Moar Production and characterization of Bacillus thuringiensis Cry1Ac-resistant cotton bollworm Helicoverpa zea (Boddie). Appl. Environ. Microbiol. 74: 462Ð469. Badenes-Perez, F., and A. M. Shelton Pest management and other agricultural practices among farmers growing cruciferous vegetables in the Central and Western Highlands of Kenya and the Western Himalayas of India. Int. J. Pest Manage. 52: 303Ð315. Bates, S. L., J. Z. Zhao, R. T. Roush, and A. M. Shelton Insect resistance management in GM crops: past, present and future. Nat. Biotechnol. 23: 57Ð62. Bhattacharya, R. C., N. Viswakarma, S. R. Bhat, P. B. Kirti, and V. L. Chopra Development of insect-resistant cabbage plants expressing a synthetic cry1ab gene from Bacillus thuringiensis. Curr. Sci. 83: 146Ð150. Brookes, G., and P. Barfoot Global impact of biotech crops: socio-economic and environmental effects, 1996Ð AgBioForum 11: 21Ð38. Cao, J., J. D. Tang, N. Strizhov, A. M. Shelton, and E. D. Earle Transgenic broccoli with high levels of Bacillus thuringiensis Cry1C protein control diamondback moth larvae resistant to Cry1A or Cry1C. Mol. Breed. 5: 131Ð 141. Cao, J., J.-Z. Zhao, T. D. Tang, A. M. Shelton, and E. D. Earle Broccoli plants with pyramided cry1ac and cry1c Bt genes control diamondback moths resistant to Cry1A and Cry1C proteins. Theor. Appl. Genet. 105: 258Ð264. Cao, J., A. M. Shelton, and E. D. Earle Development of transgenic collards (Brassica oleracea L., var. acephala) expressing a cry1ac or cry1c Bt gene for control of the diamondback moth. Crop Prot. 24: 804Ð813. Chen, M., J. Z. Zhao, H. L. Collins, E. D. Earle, J. Cao, and A. M. Shelton A critical assessment of the effects of Bt transgenic plants on parasitoids. PLoS ONE 3(5): e2284. (doi: /journal.pone ). [CIMBAA] Collaboration on Insect Management for Brassicas in Asia and Africa Collaboration on Insect

7 December 2009 SHELTON ET AL.: CRUCIFEROUS LEPIDOPTERA SUSCEPTIBILITY TO CRY1BA2 AND CRY1CA Management for Brassicas in Asia and Africa. ( [FAOSTAT] Food and Agriculture Organization of the United Nations for a world without hunger. ( faostat.fao.org/). Ferré, J., J. V. Rie, and S. C. MacIntosh Insecticidal genetically modiþed crops and insect resistance management (IRM), pp. 41Ð86. In J. Romeis, A. M. Shelton, and G. G. Kennedy [eds.], Integration of insect-resistant, genetically modiþed crops within IPM programs. Springer, Dordrecht, The Netherlands. Gould, F Sustainability of transgenic insecticidal cultivars: integrating pest genetics and ecology. Annu. Rev. Entomol. 43: 701Ð726. [HSPH] Harvard School of Public Health Fruits and vegetables. ( fruits.html). Hoy, C. W., J. E. McCully, J. A. Laborde, A. Vargas-López, R. Bujanos-Muñiz, and E. Rangel The linkage between integrated pest management and agroecosystem managementña case study in the Bajio, Mexico. Am. Entomol. 53: 174Ð183. James, C Global status of commercialized biotech/ GM Crops: ISAAA Brief No. 39, International Service for the Acquisition of Agri-Biotech Applications, Ithaca, NY. Jin, R. G., Y. B. Liu, B. E. Tabashnik, and D. Borthakur Development of transgenic cabbage (Brassica oleracea var. capitata) for insect resistance by Agrobacterium tumefaciens mediated transformation. In Vitro Cell Dev. Biol. 36: 231Ð237. Kirk, A. A., and D. Bordat Improving biological control of Plutella xylostella. CIRAD, Montpellier, France. Kovach, J., C. Petzoldt, J. Degni, and J. Tette A method to measure the environmental impact of pesticides. New YorkÕs Food and Life Sciences Bulletin 139. NYS Agricultural Experiment Station, Cornell University, Geneva, NY. Li, H., L. Bushman, F. Huang, K. Zhu, B. Bonning, and B. Oppert DiPel-selected Ostrinia nubilalis larvae are not resistant to transgenic corn expressing Bacillus thuringiensis Cry1Ab. J. Econ. Entomol. 100: 1862Ð1870. Macrae, T. C., M. E. Baur, D. J. Boethel, B. J. Fitzpatrick, A. G. Guo, J. C. Gamundi, L. A. Harrison, V. T. Kabuye, R. M. Mcpherson, J. A. Miklos, et al Laboratory and Þeld evaluations of transgenic soybean exhibiting high-dose expression of a synthetic Bacillus thuringiensis cry1a gene for control of Lepidoptera. J. Econ. Entomol. 98: 577Ð587. Metz, T. D., R. T. Roush, J. D. Tang, A. M. Shelton, and E. D. Earle Transgenic broccoli expressing a Bacillus thuringiensis insecticidal crystal protein: implications for pest resistance management strategies. Mol Breed. 1: 309Ð317. Mittal, A., A. Kumari, V. Kalia, D. K. Singh, and G. T. Gujar Spatial and temporal baseline susceptibility of diamondback moth, Plutella xylostella (Linnaeus) to Bacillus thuringiensis subsp. kurstaki spore crystal complex, puriþed crystal toxins and mixtures of Cry toxins in India. Biopestic. Int. 3: 58Ð70. Mohan, M., and G. T. Gujar Geographic variation in susceptibility of the diamondback moth, Plutella xylostella (L) (Lepidoptera: Plutellidae) to Bacillus thuringiensis spore-crystal mixtures and puriþed crystal proteins and associated resistance development in India. Bull. Entomol. Res. 92: 489Ð498. Ross, G.E.S Maximum likelihood programme. The Numerical Algorithms Gr. Rothamsted Experiment Station, Harpenden, United Kingdom. Russell, R. M., J. L. Robertson, and N. E. Savin POLO: a new computer program for probit analysis. Bull. Entomol. Soc. Am. 23: 209Ð213. Shelton, A. M., R. T. Roush, P. Wang, and J. Z. Zhao Resistance to insect pathogens and strategies to manage resistance: an update, pp. 793Ð811. In L. A. Lacey and H. K. Kaya [eds.], Field manual of techniques in invertebrate pathology, 2nd ed. Kluwer Academic, Dortrecht, The Netherlands. Shelton, A. M., M. Fuchs, and F. Shotkoski Transgenic vegetables and fruits for control of insects and insectvectored pathogens, pp. 249Ð272. In J. Romeis, A. M. Shelton, and G. G. Kennedy [eds.], Integration of insectresistant, genetically modiþed crops within IPM programs. Springer, Dordrecht, The Netherlands. Srinivasan, R., N. S. Talekar, and V. Dhawan Transgenic plants with dual Bt gene: an innovative initiative for sustainable management of Brassica insect pests. Curr. Sci. 88: 1877Ð1879. Talekar, N. S., and A. M. Shelton Biology, ecology and management of diamondback moth. Annu. Rev. Entomol. 38: 275Ð301. Wu, K., Y. Y. Guo, and N. Lv Geographic variation in susceptibility of Helicoverpa armigera (Lepidoptera: Noctuidae) to Bacillus thuringiensis insecticidal protein in China. J. Econ. Entomol. 92: 273Ð278. Zhao, J. Z., H. L. Colins, J. D. Tang, J. Cao, E. D. Earle, R. T. Roush, S. Herrero, B. Escriche, J. Ferré, and A. M. Shelton Development and characterization of diamondback moth resistance to transgenic broccoli expressing high levels of Cry1C. Appl. Environ. Microbiol. 66: 3784Ð3789. Zhao, J. Z., Y. X. Li, H. L. Collins, J. Cao, E. D. Earle, and A. M. Shelton Different cross-resistance patterns in the diamondback moth (Lepidoptera: Plutellidae) resistant to Bacillus thuringiensis toxin Cry1C. J. Econ. Entomol. 94: 1547Ð1552. Zhao, J. Z., J. Cao, Y. X. Li, H. L. Collins, R. T. Roush, E. D. Earl, and A. M. Shelton Plants expressing two Bacillus thuringiensis toxins delay insect resistance compared to single toxins used sequentially or in a mosaic. Nat. Biotechnol. 21: 1493Ð1497. Received 21 June 2009; accepted 24 August 2009.